The consequences of cloud failure
What happens to your data when the cloud crashes?
Monday, January 18 2010 || Comment || BY Rob O'Neill
Then all the other IT vendors climb on board, reposition themselves, develop their talking points, get ‘on message’, and pretty soon the media is full of stories about whatever new technology is the flavour of
the day.
The latest iteration of the phenomenon is cloud computing, a model of computing where services, processing and storage happen in a data centre somewhere far away and are accessed over the internet. The last year has been all cloud computing.
But Gartner’s hype cycle has a second and third phase — and the second, the trough of disillusionment, is the most fun.
The third, adoption, is when technologies that survive the trough get mainstreamed. This is when journalists stop writing about that technology and move on to something new.
At a recent cloud computing conference in Auckland the CEO of hosting provider OneNet, Michael Snowden, said Gartner now placed cloud computing at the very peak of the hype cycle. The trough beckoned.
Businesses of any scale have a number of reservations about the cloud, mainly in the area of security — issues such as who has access to data, where that data resides and how reliable computing over the wire really is.
Nevertheless, there is a lot of interest, and some are already shifting services such as email to a cloud model. Others are adopting software as service applications, a subset of the cloud in most definitions, such as those from local developer Xero or US giant Salesforce.com. Generally these people are having a good experience and their companies continue to grow.
But then, in early October, the cloud had its first major failure. Sure, there had been outages before, but the complete collapse of servers used to provide services to a US phone called the Sidekick, from T-Mobile, resulted in major data losses for users and the inability to activate new phones.
Those servers were operated by the aptly named Danger, which makes the Sidekick, and which Microsoft acquired in February last year.
“This goes beyond FAIL, face-palm, or any of the other internet memes we’ve come to associate with incompetence,” wrote Jason Kincaid on the influential technology blog Techcrunch. “The fact that T-Mobile and/or Microsoft Danger don’t have a redundant backup is simply inexcusable, especially given the fact that the Sidekick is totally reliant on the cloud because it doesn’t store its data locally.”
That’s right, the Sidekick is designed to be more secure, to look after users’ data better than they can themselves through the use of a supposedly robust cloud service where data gets stored and backed up professionally. Well, that was the idea, or perhaps the unique selling point.
“Regrettably, based on Microsoft/Danger’s latest recovery assessment of their systems, we must now inform you that personal information stored on your device — such as contacts, calendar entries, to-do lists or photos — that is no longer on your Sidekick almost certainly has been lost as a result of a server failure at Microsoft/Danger,” a note on the Sidekick website said after the disaster. “That said, our teams continue to work around the clock in hopes of discovering some way to recover this information. However, the likelihood of a successful outcome is extremely low.”
The case of the Sidekick illustrates how a cloud failure can have far reaching consequences. If a standard smartphone fails, only one person is affected.
But the Microsoft Danger cloud effectively aggregated data from many, many users, just as other cloud services aggregate data from many, many businesses. An outage affects not just one company, but every company using that service.
Google’s occasional outages tarnished the cloud model. The Sidekick failure has damaged it much more severely — and speeded it on its inevitable journey into the trough of disillusionment.
When you use the cloud to store such important mission critical data such as supplier and customer order history then any outage can leave you without your most important assets.
Posted by Charles at 11:19 on January 26, 2010
Murphy's Law tells you that when you need your system most- that is when it will fail. When you have your own IT management, they know when not to do sensitive upgrades etc as they know your business - the other side of the world doesn't even know what time of day it is.
Posted by Helen at 08:19 on January 18, 2010
That's a serious stuff up and worth all the heckling, however, 2 questions to put another perspective on this:
1. How many years have cloud services worked fine - any vendors out there with a clean record?
2. In comparison to cars, what effect does an IT crash have compared to a car crash... lives versus data.
Posted by Peter at 01:58 on February 3, 2010
ReplyFlag abuse